Retinal disease caused by the herpesviruses, cytomegalovirus (CMV), varicella zoster virus (VZV), and herpes simplex virus 1 (HSV-1), is the major cause of vision loss in patients with the acquired immunodeficiency syndrome (AIDS). CMV retinitis is a progressive disease which, if untreated, causes retinal detachment and blindness. It begins in the retinal capillaries and spreads directionally through the retina to the polarized retinal pigment epithelium (RPE). In cultures of polarized RPE cells, CMV preferentially infects the apical surface membrane and progeny virions spread from cell to cell across junctional complexes of the lateral membranes. Cell-cell spread is different from entry across the apical membrane, as in polarized cells these membranes contains different proteins. Preliminary experiments showed that CMV and HSV-1 mutants that lack certain glycoproteins, US8/US9 in CMV and gE/gI in HSV-1, are significantly impaired in cell-cell spread in polarized RPE cells, although they grow normally in unpolarized fibroblasts, and fail to alter tight junction proteins. This exciting finding revealed that CMV US9 and US8 are functional homologues of gE and gI in HSV-1 and VZV. HSV-1 and pseudorabies virus (PrV) gE function in pathogenesis of ocular infections by facilitating cell-cell spread of virus in the eye and in the central nervous system. CMV US8 and US9 are similar in amino acid sequence to alpha herpesvirus gE and gI and to Vibrio cholerae proteins in a virulence cassette, zonula occludens toxin (Zot) and accessory cholera entertoxin (Ace), which increase the permeability of intestinal epithelial cells. These herpesvirus and bacterial proteins are also similar to families of mammalian Ca2+ and C1- transporters, which predicts their possible functions. It is proposed to investigate the role of gE/gI homologues in pathogenesis and to elucidate their biological activities and the mechanisms by which they alter polarity of epithelial and neuronal cells. The aims are as follows. Aim 1. Determine the degree of functional similarity between CMV US9 and HSV-1 gE in altering cellular proteins that maintain polarity. Compare the properties of US9- and gE-expressing MDCK cells. Express US9 and gE in polarized RPE cells and assess changes in cell proteins that maintain polarity and integrity of the actin cytoskeleton. Evaluate effects of mutants in US9 and gE on other cell types affected in herpesvirus retinitis. Aim 2. Determine whether US9 and US8 function as a heterodimer similarly to HSV-1 gE/gI to promote cell-cell spread of virus. Study the synthesis, processing, and transport of CMV US8 and HSV- 1 gI in MDCK cells expressing these glycoproteins. Compare heterodimer formation between US8/US9 in parallel with HSV-1 gE/gI. Analyze the transport of US8/US9 and gE/gI heterodimers and their effect on patterns of cellular proteins. Evaluate the functional homology of CMV US8/US9 by constructing a HSV-1 recombinant. Aim 3. Investigate the mechanisms by which CMV US8 and US9, HSV-1 gE and gI, and their heterodimers alter the cellular proteins that maintain polarity. Determine whether US9 and gE, and US8 and gI, function as ion transporters. As an alternative, identify cellular ligands for these viral glycoproteins, as predicted experimentally by altered properties of the actin cytoskeleton. Examine the effect of site-directed mutations on the glycoprotein functions. Construct HSV-1 and CMV recombinants with mutated forms of US8/US9 and gE/gI. Findings of these studies will have far-reaching implications for understanding pathogenic mechanisms of herpesviruses and form a foundation for developing novel strategies to arrest cell-cell spread of retinal herpesvirus infections.